Latin American Capital Markets Financing the Real Economy

Transcription

1 Latin American Capital Markets Financing the Real Economy A Glass Half Empty or Half Full? Ana Carvajal Lead Financial Sector Specialist Harvard Symposium: Building the Financial System for the 21 st Century: An Agenda for Latin America and the United States December

2 KEY INDICATORS (as of 2016)* Source: WFE and FIAB, BIS and SUGEVAL, ICI and SMV, OECD, and WBG statistics (for GDP) t 2

3 A GLASS HALF EMPTY? Domestic CM have limited role in corporate financing Equity markets concentrated in a few large companies, with a limited number of new entrants More dynamism in debt markets, but still open mostly to financial and larger non-financial corporates Limited investor base Only pension funds have sizable assets across many countries in the region Life insurance at nascent stage (except for Chile) Limited direct retail investor participation and very small CIS industry (except for Brazil) Substantial foreign inflows but focusing on government bonds and larger corporates (and in many cases, financing takes place in the international markets) 3

4 OR HALF FULL? Region pioneer in mobilization of pension funds to infrastructure financing Chile (1990s-2000s) and more recently Brazil, Colombia, Mexico, Peru and even smaller countries (Costa Rica, El Salvador) Many instruments being tested (CKDs and CERPIs in Mexico, infrastructure funds in Brazil, Colombia and Peru and project bonds in all these countries) Efforts in expanding access to corporates, but impact uncertain Creation of differentiated market to attract foreign investors (Brazil) Creation of differentiated market for institutional investors (Colombia and Peru) SME markets in Brazil, Chile, Mexico and Peru (in some cases accompanied by rationalization of disclosure and CG requirements) Some innovation in products (SME guaranteed bonds in Argentina, SME receivable funds in Chile and Peru, SME bond fund in Peru, exchange platform for factoring in Chile) 4

5 WHAT HAS BEEN DIFFERENT WITH INFRASTRUCTURE FINANCING High level alignment of interests Although complex, more ability for public authorities to influence the supply side (via PPPs) Size of funding gap provides space for CM solutions to complement banking lending Infrastructure investment aligned with profile of institutional investors help to diversify portfolios and potentially also, increase profitability Understanding of challenges and sustained commitment by many public authorities to address them Efforts to identify gaps via consultations with multiple stakeholders, sometimes with MDBs (like the WBG) supporting the process Concerted policy reforms in different areas from PPP side, to capital markets and pension funds regulations and revisions to tax treatment of instruments More than one iteration of reforms Innovation in instruments with a view to aligning risk-return appetite of institutional investors (domestic and in some cases also foreign) Increasingly use of public funding to spur private sector participation (i.e. development of credit guarantees and more recently tools also to hedge FX risk with support from MBDs such as the WBG) In many cases, support from MDBs, such as the WBG, as comfort investors 5

6 WHAT HAVE WE LEARNED Sustained commitment key to accelerating mobilization of CM to corporate financing Constitute a committee with representatives from public and private sector Benchmark yourselves (see example) Make benchmark public and provide periodic updates on progress 6

7 WHAT HAVE WE LEARNED EXAMPLE OF CRITERIA TO BENCHMARK CORPORATE DEBT MARKETS (FOCUS ON DOMESTIC INVESTORS)* Access Market efficiency Aligning risk-return appetite of investors Availability of agile private offering regime Differentiated requirements for public offering depending on size of company Differentiated and efficient processes for registration of public offerings (issuance programs, seasoned issuers, etc) Tax neutrality between different types of funding (tax treatment of loans versus debt) Availability of specialized market participants to support issuance and structuring Availability of credible ratings Availability of credible research Pre and post-trade transparency aligned with market structure Alignment of market model with market structure Depth of market Availability of different types of instruments Availability of CIS structures to create volume/size and which are tax efficient Availability of credit enhancements Credible financial information from issuers Legal and regulatory environment Credible supervisor Efficient mechanisms for dispute resolution Robust but flexible insolvency regime Availability of credit information * MSCI Emerging Markets Criteria useful as a benchmark to expand foreign investors participation 7

8 WHAT HAVE WE LEARNED Innovation key to align appetite of institutional investors Use of funds to create volume to make smaller corporates more attractive to institutional investors (i.e. SME bond funds) Use of funds to link banking and real based lending into CM solutions (Funds specialized in factoring, leasing, loans, etc) Adapt current public guarantee schemes for loans to serve CM instruments (i.e. first loss in SME securitization, or SME bond funds, etc) Use innovation to expand credit information (i.e. development of credit scorings based on non traditional information) Concerted policy changes critical for true catalytic effect Supply (review of regulatory requirements to access the market, tax neutrality of different funding options, etc) Demand (investment rules of pension funds, taxation of investors) Market infrastructure (availability of reliable information, ratings, research, improvements in clearing and settlement, etc) 8

9 WHAT HAVE WE LEARNED BUT, only improvements in preconditions will lead to quantum leaps Macroeconomic conditions, size of the economy, and level of savings are key determinants of market potential, along with rule of law In general there is progress but uneven across countries in the region and with hiccups The structure of pension systems is a big ticket in the agenda with significant impact on CM development Regional integration often cited as the quick solution for smaller markets Conceptually appealing but politically complex thus take time And it does not substitute the need to work on preconditions 9

10 Annex 10

11 HOW IS THE WBG HELPING LAC IN LONG TERM FINANCE? STYLIZED TEMPLATE FOR INFRASTRUCTURE A one stop solution across the value chain, involving multiple expertise Supply side Project pipeline: Standardization, bidding procedures for capital markets solutions Project Feasibility Fund and Contingent Liability framework and training Instrument design: bonds, funds and credit enhancement instruments Demand side Long term investors rules and minimum return benchmark rules Capacity building for LT investors and regulators Sector policy changes in some cases (e.g. annuities hedging in Colombia) Improved conditions for foreign investors (e.g. taxes, fx hedging) Market architecture, infrastructure and regulations Pricing: government bond yield curve and credit rating Prudential regulations for credit enhancement instruments Market regulations: issuance, securitization, SPVs, etc. Role of MDBs (depending on context) Outright investments/interventions on balance sheet Catalytic role to crowd-in private sector Pooled financing: Bond bank models 11

12 CASE STUDY: COLOMBIA The Challenge A USD 26 bn pipeline of toll roads and other growing infrastructure needs Insufficient funding from domestic banks Shallow corporate bond market (5.6 % of GDP) Large but very concentrated pension fund industry: USD 80 billion (20 % of GDP), 4 pension funds of which 2 hold 80 % of assets. High risk averse institutional demand below AA+ 12

13 CASE STUDY: COLOMBIA The Solution: A proactive set of coordinated policies Standardization of public-private partnership (PPP) contracts Changes to infrastructure fiscal support to toll road program From 15 to 30 % availability payments in USD More frequent payments to compensate for traffic shortfalls Capital markets support Creation of a development bank, Financiera de Desarrollo Nacional (FDN), to support market-based financing solutions with emphasis on capital markets: guarantees, take out facilities New regulations for the offering of issuances to institutional investors to make the offering more efficient Changes in fund regulations to allow for infrastructure debt funds and in parallel changes to institutional investors regulations WBG support included: Supporting the infrastructure agency, Agencia Nacional de Infraestructura (ANI), in contract standardization and quality control of project structuring Investment in equity of FDN by IFC Supporting FDN in design of guarantees, standard project bonds and capacity building, Investing in infrastructure debt fund Supporting the financial supervisory agency, Superintendencia Financiera de Colombia (SFC) and the Ministry of Finance on the development of a set of regulation including regulations for the offering regime, regulations for mutual funds and project finance regulations. 13

14 The Billion Prices Project Big Data in Macroeconomics Alberto Cavallo MIT & NBER Symposium Building the Financial System Harvard Law - December 2017 Based on Cavallo & Rigobon (2016) The Billion Prices Project: Using Online Data for Measurement and Research, Journal of Economic Perspectives, 30(2)

15 Big Data in Macroeconomics Increasing demands for better data in macroeconomics (Eichengreen (2015), Summers (2016)) But macroeconomists rely mostly on data collected by governments Griliches (1985) The Uneasy Alliance Can Big Data help? (Einav & Levin (2014))

16 Big Data in Macroeconomics New technologies for data collection New not necessarily big Open access (governments cannot restrict) Traditional data sources Survey methods (National Statistical Offices) New data sources Administrative data (eg. tax, property records, banking records) Scanner data (eg. Nielsen) Search data (eg. Google, Indeed) Satellite data (eg. lights, parking lots, tanker and crop heights) Sensor data (smart phones, smart watches, IOT devices) Online data (eg. Billion Prices Project) Crowd-sourced data (eg. mobile phones)

17 The Billion Prices Project Academic initiative to collect and use online price data for economic measurement and research Daily prices since 2008 From hundreds of large multi-channel retailers In over 60 countries Use scraping technology Connect to hundreds of online retailers every day Find individual products Store key item information in a database Calculate inflation and other statistics Date Item Price Description

18 Web-Scraping Online Data Every day a software downloads a public webpage, analyses the HTML code, extract product data, and stores it in a database

19 The Case of Argentina s Missing Inflation Inflation Expectations (Survey - Di Tella University) Government starts to monitor the National Statistics Agency National Statistics Agency is officially intervened Annual Inflation Rate (%) CPI Annual Inflation CPI YoY Inflation Expectations (Di Tella University)

20 Could online prices help? Largest supermarkets posted all prices online (but few online sales at the time) Cavallo (2013) scraped daily prices for all products sold by the largest supermarket in 5 Latin American countries (about 11K daily prices each). Applied standard CPI methods : 1. Price changes at product level, 2. Unweighted geometric means within categories, 3. Aggregated using a weighted arithmetic mean and official weights for the sectors covered

21 Online indices matched CPIs in other countries Supermarket Price Index Source: Cavallo (2013) Online vs Official Price Indexes: Measuring Argentina s Inflation - Journal of Monetary Economics. Vol 60.

22 Something was wrong with Argentina... In 4 years online prices had increased 100%, the CPI only 25% Source: Cavallo (2013) Online vs Official Price Indexes: Measuring Argentina s Inflation - Journal of Monetary Economics. Vol 60.

23 Different level but similar dynamics The annual inflation rates behaved similarly over time Source: Cavallo (2013) Online vs Official Price Indexes: Measuring Argentina s Inflation - Journal of Monetary Economics. Vol 60.

24 Not explained by retailer, method, or goods Source: Cavallo (2013) Online vs Official Price Indexes: Measuring Argentina s Inflation - Journal of Monetary Economics. Vol 60.

25 Best way to match the official data? Take the online inflation rate and divide by 3. Source: Cavallo (2013) Online vs Official Price Indexes: Measuring Argentina s Inflation - Journal of Monetary Economics. Vol 60.

26 Statistical Crisis By underestimating inflation they overestimated growth

27 14 STATE STREET ASSOCIATES

28 Since 2016, Argentina s CPI closely matches our index Source: - PriceStats - State Street

29 Three related initiatives Alternative inflation indices for Argentina (online) and Venezuela (crowdsourced) Started in 2007 Academic initiative to use new data sources in macro and international research Started in Private company that collects daily price data from thousands of retailers that sell online, publishes daily inflation and PPP series (distributed by State Street) Started in

30 PriceStats Daily Inflation and PPP Indicators

31 US Daily Price Index July 2014 January 2015 January 2011 May /15/ /20/2008 Source: Cavallo & Rigobon (2016) The Billion Prices Project, Journal of Economic Perspectives, Spring 2016, Vol 30(2):

32 Anticipation: Lehman Brothers Bankruptcy Source: Cavallo & Rigobon (2016) The Billion Prices Project, Journal of Economic Perspectives, Spring 2016, Vol 30(2):

33 International Relative Prices with Online Data Illustration: International Prices The objective of the PPP series is to help determine whether currencies are under or overvalued over time Online prices represent an effective tool to measure PPP fluctuations Identical items sold around the world Detailed descriptions to achieve a nearly perfect matching Daily Prices Compare prices for a bottle of Coke across countries Apply similar approach to hundreds of products on a daily basis PPP Indices PriceStats indices: More than 300 narrow product categories With > 50 thousand individually matched items In food, fuel, and electronics High frequency releases Sector and Country-Level Indices

34 PPP Methodology Identify Define a index Product needs Select, Identify clean, size retailers to individual scrape items Scrape Product selected index retailers Repeat Run data for hundreds quality of Products check Sector Index (E.g. Food) Country Level Index (weighted) Very narrow product definition Must be available in multiple countries Branded and Unbranded categories Dozens of items per product in each country Different retailers, brands, and sizes Product availability varies across countries and time If a good is not available in the US, it will not appear in our series Some sectors are cheaper, others more expensive Use to compare Eppp and E

35 Relative Prices and Exchange Rates Real Exchange Rate The relative cost of a large basket of identical goods when expressed in the same currency =. Relative Price Level The relative price in local currencies The sub-component of the RER that we are measuring (the other is the market nominal exchange rate E) Useful to distinguish whether movements in the RER are coming from local prices or E. PPP Exchange Rates (Valuation Estimates) Use relative prices to estimate long-run expected exchange rate What is the E that takes us back to equilibrium? Answer depends on equilibrium benchmark Historical Benchmark average RER (since series started) Data can be downloaded to can create custom benchmarks (eg. recent averages or account for regime change )

36 Example: Argentina s Overvaluation (Nov. 30 th, 2017)

37 Brazil

38 We also experiment with other data sources New data sources Administrative data (eg. tax, property records, banking records) Scanner data (eg. Nielsen) Search data (eg. Google, Indeed) Satellite data (eg. lights, parking lots, tanker and crop heights) Sensor data (smart phones, smart watches, IOT devices) Online data (eg. Billion Prices Project) Crowd-sourced data (eg. mobile phones)

39 Measuring Venezuela s Inflation with Mobile Phones No official inflation data since December 2015 Crowdsourcing we hire freelancers in Venezuela to collect about 500 prices, every week, in 10 cities Mobile phones they download an app, visit a retailer, scan product barcodes, enter prices, take photo of the price tag Other freelancers/volunteers validate the prices on our website

40 Inflation is close to 60% per month Data updated every day Lower than IMF and other estimates, but rising quickly

41 Inflation Details by Category All data and code publicly available on our website

42 Final Remarks Big/New data is a great measurement opportunity Governments no longer have the monopoly of macro data We saw examples of online and crowdsourced data, but many other new technologies available Every index/database has advantages and disadvantages be aware of potential issues/biases Considering your own Big Data project? Think carefully what you want to measure Start small, experiment a lot, scale up As Griliches said, we need to get involved in the task of designing and collecting datasets of our own if we want to find the grain of relevant information in all the chaff